diff --git a/.mex/.tool-configs/.cursorrules b/.mex/.tool-configs/.cursorrules index 40ecf8f..c41ae35 100644 --- a/.mex/.tool-configs/.cursorrules +++ b/.mex/.tool-configs/.cursorrules @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/.mex/.tool-configs/.windsurfrules b/.mex/.tool-configs/.windsurfrules index 40ecf8f..c41ae35 100644 --- a/.mex/.tool-configs/.windsurfrules +++ b/.mex/.tool-configs/.windsurfrules @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/.mex/.tool-configs/CLAUDE.md b/.mex/.tool-configs/CLAUDE.md index 40ecf8f..c41ae35 100644 --- a/.mex/.tool-configs/CLAUDE.md +++ b/.mex/.tool-configs/CLAUDE.md @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/.mex/.tool-configs/copilot-instructions.md b/.mex/.tool-configs/copilot-instructions.md index 40ecf8f..c41ae35 100644 --- a/.mex/.tool-configs/copilot-instructions.md +++ b/.mex/.tool-configs/copilot-instructions.md @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/templates/.tool-configs/.cursorrules b/templates/.tool-configs/.cursorrules index 40ecf8f..c41ae35 100644 --- a/templates/.tool-configs/.cursorrules +++ b/templates/.tool-configs/.cursorrules @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/templates/.tool-configs/.windsurfrules b/templates/.tool-configs/.windsurfrules index 40ecf8f..c41ae35 100644 --- a/templates/.tool-configs/.windsurfrules +++ b/templates/.tool-configs/.windsurfrules @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/templates/.tool-configs/CLAUDE.md b/templates/.tool-configs/CLAUDE.md index 40ecf8f..c41ae35 100644 --- a/templates/.tool-configs/CLAUDE.md +++ b/templates/.tool-configs/CLAUDE.md @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/templates/.tool-configs/copilot-instructions.md b/templates/.tool-configs/copilot-instructions.md index 40ecf8f..c41ae35 100644 --- a/templates/.tool-configs/copilot-instructions.md +++ b/templates/.tool-configs/copilot-instructions.md @@ -32,10 +32,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted. diff --git a/templates/AGENTS.md b/templates/AGENTS.md index ffb5a81..62a35b1 100644 --- a/templates/AGENTS.md +++ b/templates/AGENTS.md @@ -35,10 +35,12 @@ last_updated: [YYYY-MM-DD] - Build: `npm run build` --> ## Code Graph -The repo is indexed into `.mex/graph.db`. Prefer graph commands over grepping or reading files. -- Explore a task with `mex graph scope ""` first — it returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Treat any source the graph returns as ALREADY READ; do not re-open those files. +The repo is indexed into `.mex/graph.db`. Use it to avoid re-reading code you already have — it is one tool alongside Grep/Glob, not a replacement for them. +- If you know the symbol name, go straight to it: `mex graph query ` and `mex graph get ` are exact and cheap. This is the strongest part of the graph. Give it exact names — an approximate name can return a confident wrong match. +- Exploring an unfamiliar task? `mex graph scope ""` returns a compact JSONL manifest (`meta`, `fact`s, `summary`). Scope matches on words, not meaning: if your phrasing does not share vocabulary with the code, results will be weak. Treat it as a starting point, never as a complete answer. +- If the manifest does not clearly contain what you need, use Grep/Glob instead. Do not expand node ids that look irrelevant, and do not re-run `scope` with reworded phrasing more than once — that costs more than searching directly. +- Treat any source the graph DOES return as ALREADY READ; do not re-open those files. - Pick 1-3 relevant node ids from the manifest and expand only those with `mex graph get --detail source`. -- If you already know the symbol, skip scope: use `mex graph query `, or `mex graph get `. - Before editing a symbol, run `mex impact ` to see affected callers and scaffold memory. - If a result is `truncated`, do NOT repeat the broad query — narrow the task or use the summary's `suggestedNextCommands`. Scale through a few focused calls, never one giant response. - During `mex sync`, adjudicate any AMBIGUOUS grounding; after repairs, ensure the refreshed grounding is re-emitted.